OpenMatter Network Bolsters Platform with Secure AI, Computing, and Data Collaboration Features
OpenMatter Network has unveiled significant platform enhancements, introducing MatterSDK with MatterVault for enhanced secret protection and Model Router for streamlined AI model management.
OpenMatter Network has announced a substantial expansion of its platform, less than three months after its commercial debut. The new capabilities are designed to simplify the process for enterprises, developers, and researchers to build, deploy, and collaborate using sensitive data and artificial intelligence, all while maintaining cryptographic control over information access, computation, and sharing.
These additions, now available on the OpenMatter Network platform, span secure application development, AI model management, privacy-preserving machine learning, and data collaboration. The company emphasizes that its platform is built on an extensible Verification Architecture, a cryptographic foundation that validates actions without dictating them, allowing for the integration of new technologies as enterprise computing evolves.
Key among the new features is MatterSDK, a client layer that provides developers with simplified access to MatterChain capabilities. Integrated within MatterSDK is MatterVault, which leverages threshold cryptography to safeguard secrets such as API keys and credentials. Instead of storing a complete key in one location, MatterVault distributes key shares among multiple parties, ensuring that no single machine can decrypt information independently, thereby enhancing security without requiring specialized cryptographic expertise from developers.
For organizations utilizing multiple AI providers, OpenMatter has introduced Model Router. This feature acts as a central gateway, enabling businesses to manage access to AI models from providers like OpenAI, Anthropic, and Google, as well as self-hosted on-premise endpoints. Model Router allows for the establishment of routing rules, dynamic model switching without application redeployment, centralized credential rotation, and usage monitoring, all while protecting provider keys from direct exposure in individual AI agent environments.
The platform also sees the introduction of MatterML V2, a significant advancement in privacy-preserving computing. MatterML V2 facilitates joint training and operation of AI models across combined datasets without requiring any participant to reveal their underlying data. The new version boasts substantial performance improvements, enabling secure multi-party computation through a graphical interface, thereby allowing analysts to execute complex privacy-preserving workflows without extensive coding. Early benchmarks indicate a thousand-fold increase in efficiency.
Furthermore, OpenMatter is enhancing its collaborative capabilities with the introduction of 'Communities.' These allow research groups and other member-led organizations to organize around datasets, define privacy levels, discuss information, and govern community endorsements. This aims to increase the utility of valuable data without compromising its owners' control.
CEO and Co-Founder Renee Davis highlighted that these are not isolated features but demonstrate the potential of an open Verification Architecture. The platform's design separates verification from applications, AI models, and infrastructure, enabling the seamless introduction of new capabilities while preserving a consistent cryptographic foundation for data protection and computation verification. This adaptability is crucial in the rapidly evolving AI landscape, where traditional technology refresh cycles are becoming impractical.
OpenMatter Network, headquartered in Florida, is focused on building a Verifiable Trust Layer for Secure Collaboration and AI Agents, guided by the principle "Don't Trust Data. Prove It." Their cryptographically verifiable architecture aims to enable secure collaboration, governed AI behavior, and mathematically verifiable results in untrusted environments.